Sandbox doesn't actually provide U-Boot access to the machine's physical
memory. Instead it provides a RAM buffer of configurable size, and all
memory accesses are within that buffer. Sandbox memory starts at 0 and
is CONFIG_DRAM_SIZE bytes in size. Allowing access outside this buffer
might produce unpredictable results in the event of an error, and would
expose the host machine's memory architecture to the sandbox U-Boot.
Most U-Boot functions assume that they can just access memory at given
address. For sandbox this is not true.
Add a map_sysmem() call which converts a U-Boot address to a system
address. In most cases this is a NOP, but for sandbox it returns a
pointer to that memory inside the RAM buffer.
To get a U-Boot feature to work correctly within sandbox, you should call
map_sysmem() to get a pointer to the address, and then use that address for
any U-Boot memory accesses.
Signed-off-by: Simon Glass <sjg@chromium.org>
Implements the tstc() interface for the serial driver. Multiplexing
the console between the serial port and a keyboard uses a polling
method of checking if characters are available; this means that the
serial console must be non-blocking when attempting to read
characters.
Signed-off-by: Taylor Hutt <thutt@chromium.org>
Signed-off-by: Simon Glass <sjg@chromium.org>
The buffer cannot be changed by this function, so change the buffer
pointer to a const. This allows callers with const pointer to use the
function without a cast.
Signed-off-by: Simon Glass <sjg@chromium.org>
There are two problems:
1. The argument count needs to be checked before argv is used
2. When verify is not enabled, we need to define a constant zero value
Signed-off-by: Simon Glass <sjg@chromium.org>
Add back return statement to fix compiler warning about control flow
reaching end of non void function that was introduced with:
e05e5de arm: move C runtime setup code in crt0.S
Signed-off-by: Allen Martin <amartin@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
Calculating the checksum of incompletely read data is useless.
Signed-off-by: Phil Sutter <phil.sutter@viprinet.com>
[scottwood@freescale.com: minor formatting fix]
Signed-off-by: Scott Wood <scottwood@freescale.com>
The single message is misleading, since there is no equivalent success
note when reading the other copy succeeds. Instead, warn if one of the
redundant copies could not be loaded and emphasise on the error when
reading both fails.
Signed-off-by: Phil Sutter <phil.sutter@viprinet.com>
If the NAND is locked tight, commands such as lock and unlock will not
work, but the NAND chip may not report an error. Check the lock tight
status before attempting such operations so that an error status can be
reported if we know the operation will not succeed.
Signed-off-by: Joe Hershberger <joe.hershberger@ni.com>
"nand read.part addr off size" would be treated as "nand read.raw addr off 1"
It now fails as intended stating "Unknown nand command suffix '.part'"
Signed-off-by: Harvey Chapman <hchapman@3gfp.com>
Add support for per architecture CROSS_COMPILE toolchain definitions
via CROSS_COMPILE_ARCH where "ARCH" is any of the supported u-boot
architectures. This allows building every supported u-boot board in a
single pass of MAKEALL.
Signed-off-by: Allen Martin <amartin@nvidia.com>
Acked-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Stephen Warren <swarren@wwwdotorg.org>
Currently, in gpt command, partion size is converted from string
to unsigned long type using 'ustrtol' function. That type limits
the partition size to 4GB.
This patch changes the conversion function to 'ustrtoll' to return
unsigned long long type.
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
The issue got introduced in a cleanup by Manjunath Hadli in commit
826e99136e. The eth_getenv_enetaddr_by_index
method will validate the MAC addr and if none is set in the environment
0 will be returned. Set the MAC from the eeprom if no valid address
is found in environment.
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
u-boot is not consistent if state should be 0|1 or of the enum, the
GPIO driver expects this to be one of the enum values. Update the
caller.
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
The differences include the number of GPIOs and that one is
not required to set the pinmux on request.
Signed-off-by: Holger Hans Peter Freyther <holger@freyther.de>
Add 'ustrtoull' function to convert size from string (ex: 1GiB)
to unsigned long long type
Signed-off-by: Piotr Wilczek <p.wilczek@samsung.com>
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
On architectures where manual relocation
is needed, the 'malloc_bin_reloc' function
must be called after 'mem_malloc_init'.
Make the 'malloc_bin_reloc' function static
and call it directly from 'mem_malloc_init'
instead of calling that from board_init_{r,f}
functions of the affected architectures.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Cc: Wolfgang Denk <wd@denx.de>
Cc: Andreas Bießmann <andreas.devel@gmail.com>
Cc: Jason Jin <Jason.jin@freescale.com>
Cc: Macpaul Lin <macpaul@andestech.com>
Cc: Daniel Hellstrom <daniel@gaisler.com>
Cc: Daniel Schwierzeck <daniel.schwierzeck@googlemail.com>
This is available on other architectures, and nds32 will start to break
without it as code starts to use error numbers more.
Signed-off-by: Simon Glass <sjg@chromium.org>
Remove it from the processor specific headers. This is
already defined in the common header km83xx.h.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
This board is similar to TUXX1 but it has a different sized second
FPGA. Therefore the configuration for the third chipselect is different.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
This is a preparation for the upcoming kmopti2 board. This board has
also a second fpga on board which is different to the tuxx1 target. But we
want to use the same header file. So remove the config option
KM_DISABLE_APP2 and simply use the board names to distinguish the features.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
If CONFIG_NAND_ECC_BCH is chosen from in the board configuration we add
an ecc mode to the kernel commandline.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
If CONFIG_NAND_ECC_BCH is set we use 4-bit error corretion code
instead of the 1-bit error correction code on the NAND device
within this driver.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Acked-by: Scott Wood <scottwood@freescale.com>
Replace uImage with ${uimage}.
If uimage is not set, default it to uImage.
Signed-off-by: Andreas Huber <andreas.huber@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Replace the hardcoded string with a variable. If CONFIG_NAND_ECC_BCH is
set we use a specific name for the uImage (ecc_bch_uImage).
Signed-off-by: Andreas Huber <andreas.huber@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
kmvect1 has a UEC2 connection to the piggy board and a UEC0 connection
to the switch MV88E6122. This switch has a connection to a frontport
ethernet interface. The ethernet port used for network booting is
automatically selected by u-boot. If a Piggy is plugged, the Piggy
port is selected (UEC2, eth1). If the Piggy isn't present, the
Frontport is selected (UEC0, eth0).
The switch reset is connected to a GPIO on the PRIO3 board FPGA (GPIO28)
and released at startup.
Signed-off-by: Karlheinz Jerg <karlheinz.jerg@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
For the the kmvect1 board we will also need a functionality to add an
offset to the IVMs MAC address, because these board will have two valid
ethernet ports for debugging purpose. So move the code to an own
function.
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
For kmvect1 we need a special solution and for km_arm boards we already
have. So move the common code to the architectur specific file.
Signed-off-by: Karlheinz Jerg <karlheinz.jerg@keymile.com>
Signed-off-by: Holger Brunck <holger.brunck@keymile.com>
Remove the manual relocation of env_name_spec. This has been missed
in the previous patch series for introducing dynamic relocation
on MIPS.
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
This code is pretty old and we want to support only 32-bit systems now.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Graeme Russ <graeme.russ@gmail.com>
Invert the polarity of this option to simplify the Makefile logic.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Gabe Black <gabeblack@chromium.org>
These lines are dealt with in the x86 Makefile and link script, so punt
them.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Gabe Black <gabeblack@chromium.org>
This x86 CPU variant is no longer required as the boards that use it have
been removed.
Signed-off-by: Simon Glass <sjg@chromium.org>
Acked-by: Graeme Russ <graeme.russ@gmail.com>
The code handles relocation entries with the
following relocation types only:
mips32: R_MIPS_REL32
mips64: R_MIPS_REL+R_MIPS_64
xburst: R_MIPS_REL32
Other relocation entries are skipped without
processing. The code must be extended if other
relocation types must be supported.
Add -pie to LDFLAGS_FINAL to generate the .rel.dyn
fixup table, which will be applied to the relocated
image before transferring control to it.
The CONFIG_NEEDS_MANUAL_RELOC is not needed
after the patch, so remove that as well.
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
Signed-off-by: Daniel Schwierzeck <daniel.schwierzeck@gmail.com>
Simon Glass